Run WizardLM 33B v1.0 on Mac
Uncensored conversational model based on Llama 33B with reduced refusal bias
WizardLM 33B v1.0 runs 100% private on Mac inside Private LLM — no internet connection required, no data sent to any server.
Model on Hugging Face
Specifications
- Parameters
- 33B
- Context window
- 2K tokens
- License
- Other
- Quantization
- OmniQuant 4-bit
- Family
- Llama 33B
What WizardLM 33B v1.0 is good at
WizardLM 33B v1.0 is a retrained version of WizardLM 30B that was built to be less prone to refusing or dodging questions. It comes from the Llama 33B model family, has a 2048-token context window, and was fine-tuned on a filtered dataset to lower bias and avoidance. The model acts more compliant than an earlier WizardLM 7B release and uses Vicuna-style prompt formatting. It has fewer built-in restraints, so it can generate responses on sensitive or unrestricted topics, but it still carries the underlying ethical leanings of the original Llama base. Users are fully responsible for any output they publish or act on.
Which of your devices can run it
Mac
How to run WizardLM 33B v1.0 in Private LLM
- Download Private LLM from the App Store.
- Open the in-app model library and choose WizardLM 33B v1.0.
- Download the model once, then chat fully offline.
Variants & related models
Amoral Gemma 3 1B v2
Uncensored Gemma 3 1B finetune for neutral factual responses
DeepSeek R1 Distill Llama 8B Abliterated
Uncensored DeepSeek R1 Distill 8B via refusal abliteration
DeepSeek R1 Distill Qwen 32B Abliterated
Uncensored proof-of-concept removing alignment guardrails from DeepSeek R1 Distill
Dolphin 2.6 Phi 2
Uncensored conversation, roleplay, and agent tasks from a Phi 2 base
Dolphin 2.6 Mixtral 8x7B
Uncensored Mixtral 8x7B finetune excelling at coding tasks
Frequently asked questions
WizardLM 33B v1.0 is too large for current iPhones. It runs on Mac with enough unified memory inside Private LLM, fully offline.
Yes. WizardLM 33B v1.0 runs on Macs with enough unified memory, such as Mac (Apple Silicon, 192GB), MacBook Pro (M4 Max, 128GB), Mac Studio / Pro (Apple Silicon, 96GB), MacBook Pro (M4 Max, 64GB), MacBook Pro (M4 Max, 48GB), MacBook Pro (M4 Max, 36GB), Mac (Apple Silicon, 32GB), MacBook Air (M4, 24GB), fully on-device in Private LLM.
Yes. Once downloaded in Private LLM, WizardLM 33B v1.0 runs 100% on-device — no internet connection, and nothing is sent to any server.
Private LLM is a one-time purchase with no subscription and no per-message cost. The models themselves are open source — once downloaded, they run offline with nothing to pay per use.
Why run WizardLM 33B v1.0 in Private LLM
Private LLM has run local AI on iPhone, iPad, and Mac since 2023, well before Apple Intelligence, LM Studio, Ollama, etc. existed. Inference happens on your device, so your WizardLM 33B v1.0 conversations never reach a server. The part most apps gloss over is quantization, and that is exactly where on-device quality is won or lost. Most llama.cpp and MLX wrappers ship the same off-the-shelf 4-bit RTN weights. Private LLM ships GPTQ and OmniQuant quantization, tuned per model, and our 3-bit OmniQuant models match or beat those 4-bit RTN builds on the same Apple Silicon. Run the same model both ways and you feel it in the first reply. See how our quantization works.
Specifications and summary come from WizardLM 33B v1.0's Hugging Face model card. Private LLM ships its own quantized models, built with OmniQuant quantization tuned per model, and isn't affiliated with the model's authors.